Abstract P2-06-05: p27 Expression in Hormone Sensitive Breast Carcinomas, Prior and after Short Term Hormone Treatment: A Prospective Placebo Double-Blind Study

Background: Mutations in genes that regulate the cell cycle are found in most human cancers. In addition, there are other cell cycle genes, including those encoding p27, that contribute to tumor progression but are rarely mutated. p27 is a member of the Cip/Kip family of cyclin-dependent kinase inhibitors and it is present in high levels in quiescent cells, but its levels decline when cells proliferate in response to mitogenic signals. Various functions have been attributed to p27, including promotion of apoptosis and regulation of drug resistance. p27 is functionally inactivated in most of human cancers through enhanced p27 proteolysis, sequestration by cyclin D/cdk complexes, and by cytoplasmic mislocalization. There is evidence from preclinical studies that p27 is essential for cell cycle arrest by tamoxifen and other antiestrogenic drugs.

Objectives: To compare the expression of nuclear and cytoplasmic p27, cyclin D1, Ki-67, p53, HER2/ErbB2, ER and PR in hormone sensitive invasive ductal carcinomas (IDC) in postmenopausal women, prior and after tamoxifen and anastrozole in a short term treatment. Methods: Seventy one patients were enrolled in this study and fifty-eight were elegible for analyses. All patients had palpable ER-positive IDC (stage II and III) and were double-blind randomized in a prospective placebo controlled study with short term hormone treatment (HT) groups for twenty-six days prior to surgical treatment: Anastrozole 1mg/day (n= 18), Placebo (n=25) and Tamoxifen 20mg/day (n= 15). Pre and post HT samples were disposed in tissue micro array blocks and submitted to immunohistochemical assay. Biomarkers status (nuclear and cytoplasmic p27, cyclin D1, Ki-67, p53, HER2/ErbB2, ER and PR) were obtained comparing each immunohistochemical evaluation of pre and post-surgery samples using semi-quantitative Allred's method for nuclear biomarkers, and scored based on a four-point intensity based system: negative (no stain), weakly positive (1+), mild positive (2+) and strongly positive (3+). Statistical analysis were performed using the GEE (General Estimation Equations) and ANOVA tests with significant p ≥0.05. Results: The patients mean age was 66.25 ± 9.95 years. The mean tumor size was 4.0 ± 1.06 cm. There was no statistically significant correlation between nuclear p27 expression and tumor size or histologic grade. Before HT, 38 (62.5%) patients had positive nuclear p27 expression, 5 (8.6%) patients had positive cytoplasmic p27 expression, and after treatment their expression was increased in all groups. There was a positive correlation between Ki-67, cyclin D1, p53 and nuclear p27 prior to HT. After HT, there was a positive correlation between Ki-67, cyclin D1, p53, HER2/ErbB2, ER, and nuclear p27.

Conclusions: There was an increment of nuclear and cytoplasmic p27 after short term treatment, regardless of the drug used. The positive correlation between nuclear p27 and poor prognostic markers was unexpected.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-06-05.